TEXT

Top Abstract Text References

Halogenated organic chemicals have been introduced into the environment from a variety of sources, including the improper disposal of degreasers and solvents (14). Many of the halogenated compounds, such as perchloroethylene (PCE) and carbon tetrachloride, are persistent in the environment due to their resistance to microbial degradation and their toxicity to microorganisms. For the treatment of these recalcitrant compounds, a number of chemical processes have been investigated, including oxidation by Fenton's reagent. Compound degradation by Fenton's reagent may yield (i) partial mineralization (15), (ii) lowered toxicity (1), and (iii) increased susceptibility to biodegradation (4). Owing to their ability to lower the toxicity and increase the biodegradability of the parent compounds, chemical reactions have been coupled in sequence with biological reactions and have been the subject of much recent research. An extensive review is available elsewhere (19). However, the investigation of simultaneous chemical and biological transformation processes has received little attention. Simultaneous processes could have both economic and process advantages if applied to industrial pollution or in situ hazardous waste treatment in the environment.

Fenton (8) discovered the strong oxidizing power of mixtures of hydrogen peroxide and ferrous iron solutions. Haber and Weiss (10) later identified the oxidizing element in Fenton's reagent as a hydroxyl radical. The hydroxyl radical is a nonspecific, strong oxidant which reacts with most organic and biological molecules at near diffusion-controlled rates (>10⁹ M¹ s¹) (7, 12). The classic Fenton's reaction involves the addition of dilute hydrogen peroxide to a degassed, acidic ferrous iron solution, which generates hydroxyl radicals (equation 1). The degradation of organic chemicals by hydroxyl radicals then proceeds via hydroxylation, hydrogen atom abstraction, or dimerization (21).

(1)

Some environmental applications of Fenton's reagent involve reaction modifications, including the use of high concentrations of hydrogen peroxide, the substitution of different catalysts such as ferric iron and naturally occurring iron oxides, and the use of phosphate-buffered media and metal-chelating agents. These conditions, although not as stoichiometrically efficient as the standard Fenton's reactions, are often necessary to treat industrial waste streams and contaminants in soils and groundwater (20).

As a part of our overall simultaneous chemical and biological transformation study, PCE was chosen as a probe compound for experimental investigation. PCE was known to be degraded by Fenton's reagent, yielding dichloroacetic acid (DCAA) (14, 18), a readily biodegradable compound. A major obstacle to combining the chemical and microbial reactions simultaneously was the toxicity of Fenton's reagent to microorganisms. We previously showed that the toxic effects of Fenton's reagent were sufficiently reduced by preacclimation of microorganisms to high concentrations of hydrogen peroxide, thereby allowing a significant number of microorganisms to survive throughout the course of treatment (3). In a later study, we investigated coexisting chemical and biological reactions used for mineralization of PCE (2). The results of that study, with ¹⁴C-labeled PCE, showed that the addition of microorganisms increased the extent of mineralization of PCE by more than 10% over that of a noninoculated control. This finding suggested that chemical and biological reactions could coexist and might be a viable alternative for the treatment of wastewaters containing PCE. In the present study, we investigated the effects of variables of coexistent chemical and biological reactions (concentrations of PCE, hydrogen peroxide, and ferrous iron and the initial cell number) on mineralization of PCE.

Chemicals. PCE and DCAA-sodium salt were purchased from Aldrich Chemical Co. (Milwaukee, Wis.); ¹⁴C[1-2]-PCE was obtained from Sigma Chemical Co. (St. Louis, Mo.); hydrogen peroxide (30%) was purchased from Fisher Scientific (Fair Lawn, N.J.); and Ecolite scintillation cocktail was purchased from ICN Pharmaceuticals (Costa Mesa, Calif.). Carbo-Sorb was obtained from Packard Instruments (Meriden, Conn.), and Ready Organic was purchased from Beckman (Fullerton, Calif.).

Organism and culture conditions. A hydrogen peroxide-resistant strain of Xanthobacter flavus FB71 (3) was used for the biotransformation of DCAA, the major product of PCE degradation by modified Fenton's reagent (14, 18). Cells were grown in a continuous-flow fermentor (New Brunswick Scientific BioFlo III) in a 1.5-liter volume, at steady state, with growth medium replacement at a rate of 1 liter per day at 30°C. The fermentor was stirred at 200 rpm and aerated with 3 liters of sterile air flow per min. One liter of medium contained 300 mg of DCAA per liter, 200 mg of H₂O₂ per liter, 3.4 g of Na₂HPO₄, 1.5 g of KH₂PO₄, 0.25 g of NaCl, 0.5 g of NH₄Cl, 0.12 g of MgSO₄, 5.55 mg of CaCl₂, 2 ml of Wolfe's mineral solution (9), and 2 ml of vitamin supplement solution containing 50 mg each of biotin, thiamine HCl, and nicotinic acid per liter.

Experimental design. In order to decrease the number of experiments while maintaining statistical significance, a central composite, rotatable experimental design was developed as outlined by Cochran and Cox (5). The four-level design, shown in Table 1, included PCE, H₂O₂, and Fe(II)-chelate concentrations and the initial cell number as the experimental variables. The percent mineralization, determined by radiotracer analysis, was the response measured. In addition, microbial survival at the end of the each experimental point was surveyed qualitatively by using nutrient agar plates and the spread plate technique. The design contained three blocks of experiments based on a second-order design. Block 1 was a 2³ factorial design that formed the first-order portion of the design. Block 2 provided the "star" points, which provided the second-order portion of the design, and block 3 was defined as the center of the experimental design, providing an estimate of the error of the measurements.

Experimental setup and ¹⁴C radiotracer experiments. Experiments were performed according to the experimental design outlined in Table 1 and according to the procedures given below. In order to minimize the autooxidation of ferrous iron and the loss of PCE by volatilization, the experimental order was as follows: (i) 55-ml serum bottles were filled with M9 buffer solution and distilled water so that the final solution would result in half-strength M9 buffer solution in 40 ml; (ii) hydrogen peroxide was added; (iii) microorganisms were added (except for controls); (iv) bottles were crimped; (v) PCE was added with a syringe to a resulting concentration of 5 mg/liter; and (vi) reactions were initiated by the addition of Fe(II)-nitrilotriacetic acid (NTA) solution. (Ferrous iron solution [10 mM] was prepared anaerobically by using 100 mM NTA solution degassed for 15 min with a 10-ml/min flow each of nitrogen and hydrogen.) The bottles were then incubated on a rotary platform shaker at 200 rpm at an ambient temperature for 72 h. Hydrogen peroxide and ferrous iron solutions were added every hour as indicated in Table 1 for the first 5 h of the experiments.

View larger version (17K): [in this window] [in a new window]   FIG. 1.   14C purge and trap apparatus.

Results and discussion. The results of the experiments, the observed percent PCE mineralization, along with the experimental variable values, are shown in Table 1. A statistical analysis software package (JMP version 3.2, professional edition; SAS Institute, Inc., Cary, N.C.) was used to analyze the results, provide estimated mineralization values based on a second-order model fit to the data (Table 1), and generate response surface plots. A second-order model, typical for response surface equations (5), was reduced to the following form based on the results of this experimentation:

(2)

where I is the intercept, C is the initial cell number (log cell number), H is the hydrogen peroxide concentration (in milligrams per liter), F is the ferrous iron concentration (millimolar), and a, b, c, d, e, and f are the coefficients of the experimental variables. Coefficients of the second-order model variables and the corresponding standard errors are given in Table 2. All model terms were analyzed by t test, and any term found to be insignificant at  = 0.05 was excluded from the model. Although PCE was set as a variable in the preparation of the model, the statistical analysis showed that the concentration of PCE, within the tested boundaries of the model, had no significant effect on the model. In addition, all cross terms (e.g., F × C) were found to be statistically insignificant. It should be noted that the experimental points 1, 2, and 3 from Table 1 had to be rerun due to procedural problems that occurred during the original experimental run. Three control experiments on the center location (Table 1) were conducted in parallel with the repeated points to check for their validity. The mean and standard error values of additional controls (data not shown) were within the range of the original experiments, leading to the conclusion that the rerun points were valid.

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Regression plot of observed data against predicted values from the second-order response surface model describing simultaneous chemical and biological mineralization of PCE.

View larger version (25K): [in this window] [in a new window]   FIG. 3.   Response surface plot for simultaneous chemical and biological mineralization of PCE at a ferrous iron concentration of ca. 0.3 mM.